chunked/nltk_chunking/model_doc_owlvit/chunk_8.txt

We provide the adaptation strategies and regularizations needed to attain very strong performance on zero-shot text-conditioned and one-shot image-conditioned object detection.